Cleaning up order of include files by sorting them alphabetically
keeping in mind to leave common.h on top.
Signed-off-by: Marcel Ziswiler <marcel@ziswiler.com>
Cleaning up order of include files by sorting them alphabetically
keeping in mind to leave common.h on top.
Signed-off-by: Marcel Ziswiler <marcel@ziswiler.com>
Cleaning up order of include files by sorting them alphabetically
keeping in mind to leave common.h on top.
Signed-off-by: Marcel Ziswiler <marcel@ziswiler.com>
This patch moves pxa to the common timer functions added in commit
8dfafdd - Introduce common timer functions <Rob Herring>
The (removed) pxa timer code (specifically __udelay()) could deadlock at
the 32-bit boundary of get_ticks(). get_ticks() returned a 32-bit value
cast up to a 64-bit value. If get_ticks() + tmo in __udelay() crossed
the 32-bit boundary, the while condition became unconditionally true and
locked the processor. Rather than patch the specific pxa issues, simply
move everything over to the common code.
Signed-off-by: Andrew Ruder <andrew.ruder@elecsyscorp.com>
Cc: Marek Vasut <marex@denx.de>
Cc: Tom Rini <trini@konsuiko.com>
Since commit 3ff46cc42b ("arm: relocate the exception vectors")
pxa does not boot anymore.
Add a specific relocate_vectors macro that skips the vector relocation,
as the pxa SoC does not provide RAM at the high vectors address
(0xFFFF0000), and (0x00000000) maps to ROM.
This allows pxa to boot again.
Cc: Marek Vasut <marex@denx.de>
Signed-off-by: Andrew Ruder <andrew.ruder@elecsyscorp.com>
Conflicts:
configs/socfpga_arria5_defconfig
configs/socfpga_cyclone5_defconfig
configs/socfpga_socrates_defconfig
Merged these by hand and re-ran savedefconfig on them.
Signed-off-by: Tom Rini <trini@konsulko.com>
Pull out the ArriaV SoCDK clock config from ancient Altera U-Boot
"rel_socfpga_v2013.01.01_15.05.01_pr" and implant those values into
mainline to get a booting ArriaV SoCDK.
Signed-off-by: Marek Vasut <marex@denx.de>
Synchronise the config options with Cyclone V SoCDK and other boards.
This enables ethernet on the ArriaV SoCDK.
Signed-off-by: Marek Vasut <marex@denx.de>
Add the missing DT nodes, so that ArriaV SoCDK can boot from SD
card. The SD card must be in slot J5 and BSEL must be 0x5.
Signed-off-by: Marek Vasut <marex@denx.de>
Enable the DWAPB GPIO driver for SoCFPGA Cyclone V and Arria V.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Simon Glass <sjg@chromium.org>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>
Add "bank-name" property to each GPIO bank to give it unique name.
The approach here is exactly the same as with the "regulator-name"
property for regulators.
Signed-off-by: Marek Vasut <marex@denx.de>
Add driver for the DesignWare APB GPIO IP block.
This driver is DM capable and probes from DT.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Simon Glass <sjg@chromium.org>
Now that we're actually converting the QTS-generated header files,
we can even adjust their data types. A good candidate for this is
the pinmux table, where each entry can have value in the range of
0..3, but each element is declared as unsigned long. By changing
the type to u8, we can save over 600 Bytes from the SPL, so do it.
This patch also constifies the array.
Signed-off-by: Marek Vasut <marex@denx.de>
Add script which loads the QTS-generated sources and headers and converts
them into sensible format which can be used with much more easy in mainline
U-Boot. The script also filters out macros which makes no sense anymore, so
they don't pollute namespace and waste space.
Signed-off-by: Marek Vasut <marex@denx.de>
The board/altera/socfpga directory is not a generic SoCFPGA machine
anymore, but instead it represents the Altera SoCDK board. To make
matters more complicated, it represents both CycloneV and ArriaV
variant.
On the other hand, nowadays, the content of this board directory is
mostly comprised of QTS-generated header files, while all the generic
code is in arch/arm/mach-socfpga already.
Thus, this patch splits the board/altera/socfpga into a separate
board directory for ArriaV SoCDK and CycloneV SoCDK, so that each
can be populated with the correct QTS-generated header files for
that particular board.
Signed-off-by: Marek Vasut <marex@denx.de>
The CONFIG_TARGET_SOCFPGA_CYCLONE5 and CONFIG_TARGET_SOCFPGA_ARRIA5
selected both a board and a CPU. This is not correct as these macros
are supposed to select only board.
All would be good, if QTS-generated header files didn't check for
these macros exactly to determine if the platform is Cyclone V or
Arria V. Thus, for the sake of compatibility with not well fleshed
out header file generator, this patch makes these two macros into
a stub config option and introduces new CONFIG_TARGET_SOCFPGA_CYCLONE5_SOCDK
and CONFIG_TARGET_SOCFPGA_ARRIA5_SOCDK targets, which select the
previous stub config option.
The result is that compatibility with QTS is preserved and the new
CONFIG_TARGET_* select actual target boards.
Signed-off-by: Marek Vasut <marex@denx.de>
Move the wrappers for QTS-generated files into platform directory
out of the board directory. The trick here is to add -I to CFLAGS
such that it points to the board directory in source tree and thus
the qts/ directory there is still reachable.
Signed-off-by: Marek Vasut <marex@denx.de>
The GMAC which is enabled is purely board property, so do not enable
arbitrary GMAC in DT include files. Same goes for PHY mode, which is
again a board property. The CycloneV SoCDK does this correctly, but
SoCrates doesn't. This bug never manifested itself though, since all
the boards ever used the GMAC1 . This bug manifests itself only on
boards that utilise GMAC0.
Signed-off-by: Marek Vasut <marex@denx.de>
The socfpga_cyclone5.dtsi has an mmc0 node, socrates has mmc node.
This makes aliases not very usable, so make everything into mmc0.
Moreover, zap the useless mmc alias while at this.
Signed-off-by: Marek Vasut <marex@denx.de>
This code claims it needs to wait 7us, yet it uses get_timer() function
which operates with millisecond granularity. Use timer_get_us() instead,
which operates with microsecond granularity.
Signed-off-by: Marek Vasut <marex@denx.de>
Based on observation, this udelay(20) was apparently too high and caused
subsequent failure to calibrate DDR when U-Boot was compiled with certain
toolchains. Lowering this delay fixed the problem.
Instead of permanently lowering the delay, calculate the correct delay
based on the original comment, that is, obtain EOSC1 frequency and use
it to calculate the precise delay.
Signed-off-by: Marek Vasut <marex@denx.de>
Fix the following problem:
drivers/ddr/altera/sequencer.c: In function 'sdram_calibration_full':
drivers/ddr/altera/sequencer.c:1943:25: warning: 'found_failing_read' may be used uninitialized in this function [-Wmaybe-uninitialized]
if (found_passing_read && found_failing_read)
^
drivers/ddr/altera/sequencer.c:1803:26: note: 'found_failing_read' was declared here
u32 found_passing_read, found_failing_read, initial_failing_dtap;
^
Signed-off-by: Marek Vasut <marex@denx.de>
This gem is really really rare, there was an actual float used in
the Altera DDR init code, which pulled in floating point ops from
the libgcc, just wow.
Since we don't support floating point operations the same way Linux
does not support them, replace this with an integer multiplication
and division combo. This removes some 2kiB of size from the SPL as
the floating point ops are no longer pulled in from libgcc.
This was detected by enabling CONFIG_USE_PRIVATE_LIBGCC=y , which
does not contain the floating point bits.
Signed-off-by: Marek Vasut <marex@denx.de>
Move config for the E1000 Ethernet driver to Kconfig and tidy up affected
boards.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Several files are out of order. This means that when the moveconfig tool
moves CONFIG options to Kconfig it generates a large diff. To avoid this,
reorder the files first.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add Kconfig options in preparation for moving boards to use Kconfig.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Update this driver to support driver model.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Tested-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Tested-on: Apalis T30 2GB on Apalis Evaluation Board
Since struct eth_device does not exist with CONFIG_DM_ETH defined, avoid
using it in the driver unless necessary. Most of the time it is better to
pass the private driver pointer anyway.
Also refactor the code so that code that the driver model implementation
will share are available in functions that can be called. Add stubs where
necessary.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Tested-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Tested-on: Apalis T30 2GB on Apalis Evaluation Board
We cannot currently include any header files in the C files since common.h
needs to be included first, and it is in the header file. Move it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Tested-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Tested-on: Apalis T30 2GB on Apalis Evaluation Board
Current behavior is that if CTRL+C is pressed command returns 0 that was
successful which is not correct behavior.
The easiest test case is "tftpboot 80000 uImage && echo yes"
and press CTRL+C. Then the second command is called which is incorrect.
Error log:
zynq-uboot> tftpb 80000 uImage && echo yes
Gem.e000b000:7 is connected to Gem.e000b000. Reconnecting to
Gem.e000b000
Gem.e000b000 Waiting for PHY auto negotiation to complete....... done
Using Gem.e000b000 device
TFTP from server 192.168.0.102; our IP address is 192.168.0.101
Filename 'uImage'.
Load address: 0x80000
Loading: ################
Abort
yes
zynq-uboot>
This patch adds -EINTR return value when CTRL+C is pressed.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
High 32-bit address is needed when u-boot runs in 64-bit space.
Tested on armv8-based LS2085ARDB.
Signed-off-by: Mingkai Hu <Mingkai.Hu@freescale.com>
Signed-off-by: York Sun <yorksun@freescale.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Adjust timouts and retry counts to be suitable for loaded ethernet
network. With 5 seconds timeout, 10 retries maximum, tftp is
impossible even on local network with single full-speed TCP
connection.
100msec timeout should be suitable for most networks tftp is used on,
that is local ethernets. Timeout count really needs to be way higher,
as lost packets are normal when TCP is running over the same network.
Enforce 10msec minimum.
Signed-off-by: Pavel Machek <pavel@denx.de>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Also move the spi flash configurations to the at91-sama5_common.h.
Current at91 zImage size is about 3.3M, the old mapping is not
suitable. So update the spi flash map as following:
0x0 ~ 0x004000: at91bootstrap(16k)
0x04000 ~ 0x008000: u-boot env(16k)
0x08000 ~ 0x060000: u-boot(352k)
0x60000 ~ 0x06c000: dtb (48k)
0x6c000 ~ 0x400000: kernel (3M+592k)
In AT91Bootstrap, the U-Boot in spi flash also update to 0x8000, refer
to following commit in AT91Bootstrap:
3e91e54 Kconfig: fix spi flash address
So also update SPL's u-boot load address to 0x8000 in spi flash.
Signed-off-by: Josh Wu <josh.wu@atmel.com>
Reviewed-by: Jagan Teki <jteki@openedev.com>
Since sama5d3xek boards has different type of dtb blobs, so we need to detect
the cpu type in runtime.
So we add a new variable $dtb_name. if $dtb_name is not defined, we just use
at91-${board_name}.dtb
as the $dtb_name. Otherwise, we will just load the dtb with
$dtb_name.
For sama5d3xek, we will detect cpu type and make up $dtb_name in
runtime.
Signed-off-by: Josh Wu <josh.wu@atmel.com>
As almost all sama5 sd/mmc env configurations are same, so move them to
at91-sama5_common.h.
Also define CONFIG_ENV_VARS_UBOOT_CONFIG to have the varaible: $board_name.
Then we can use 'at91-${board_name}.dtb' as the dtb name.
TODO: since sama5d3xek has different dtb name, we need to some extra
stuff to make it work on sama5d3xek boards.
Signed-off-by: Josh Wu <josh.wu@atmel.com>
To enable the clocks on the at91 boards a constant (0x4) is used.
This is replaced with a define in at91_pmc.h (1 << 2).
Signed-off-by: Erik van Luijk <evanluijk@interact.nl>
Reviewed-by: Andreas Bießmann <andreas.devel@googlemail.com>
On these boards the DDR is connected to a dedicated controller and not
to chip select 1 of the EBI.
Signed-off-by: Erik van Luijk <evanluijk@interact.nl>
Tested-by: Erik van Luijk <evanluijk@interact.nl>
The mpddr.c depends on ATMEL_BASE_MPDDRC for the base address to configure the controller.
This cannot be used when there is more than one controller (i.e. AT91SAM9G45, AT91SAM9M10).
Signed-off-by: Erik van Luijk <evanluijk@interact.nl>
[remove 'new blank line at EOF']
Signed-off-by: Andreas Bießmann <andreas.devel@googlemail.com>
The syntax for the fdt_chosen/initrd
functions seem to deprecated in usage
from MIPS bootm implementation.
Third parameter is no more used in these api's
Refer to : include/fdt_support.h
Signed-off-by: Govindraj Raja <govindraj.raja@imgtec.com>
In commit a18a477 (MIPS: use common code from lib/time.c) MIPS platforms
started using common the common timer functions which are based around
the fact that many platforms have a 32-bit free running counter register
that can be used see commit 8dfafdd (Introduce common timer functions).
Even MIPS64 has such a 32-bit register (some have an additional 64-bit free
running counter, but that's something for another time).
The problem is that in __read_32bit_c0_register() we read the value from
this register into an _signed_ int and as it's returned up the call
chain to timer_read_counter() it gets assigned to an unsigned long. On a
32-bit system there is no problem. On a 64-bit system odd things happen,
sign extension seems to kick in and all of a sudden if the counter
register happens to have the MSb (i.e. the sign bit) set the negative
int gets sign extended into a very large unsigned long value. This in
turn throws out things from get_ticks() up.
Update __read_32bit_c0_register() and __read_32bit_c0_ctrl_register() to
use "unsigned int res;" instead of "int res;". There seems to be little
reason to treat these register values as signed. They are either
counters (which by definition are unsigned) or are made up of various
bit fields to be interpreted as per the CPU datasheet.
Reported-by: Sachin Surendran <sachin.surendran@alliedtelesis.co.nz>
Signed-off-by: Chris Packham <judge.packham@gmail.com>
This has not been converted to Generic Board, so should be removed.
(See doc/README.generic-board for details.)
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Acked-by: Heiko Schocher <hs@denx.de>
Acked-by: Marek Vasut <marex@denx.de>
Marcel Ziswiler
Andrew Ruder
Tom Rini
Marek Vasut
Simon Glass
Michal Simek
Mingkai Hu
Pavel Machek
Wu, Josh
Erik van Luijk
Govindraj Raja
Chris Packham
Masahiro Yamada